Machine Learning -Level 1 Module 1 - Oumaima

Overview of things learned :

Technical Area:

  • Getting familiar with the workflow of a ML project :
    1. Understanding the business probelm
    2. Collecting data
    3. preprocessing data
    4. Data exploration
    5. ML models
    6. Models evaluation
    7. Results

  • Discovering the recommandation system and its types ( collaboratives filtering, Content based filtering) and measures of similarity ( Cosine, Dot Product and Euclidean distance)

  • Discovering NLP, and the importance of converting our data (text) to a numeric values so our machine can learn from it, for that we can use several representation like: One Hot Encoding, Bag of Words, TF-IDF, Word embedding.

  • Discovering how we can gather our data using web scraping (getting data from a single page) and web crawling (getting data from multiple pages), but we have make sure that the website allows the scrapping/crawling.

Tools:

  • BeautifulSoup, Scrapy and Selenium for web scrapping/crawling.
  • Pytorch for deep learning algorithms
  • Getting familiar with google Colaboratory
  • Spacy for NLP
  • Trello a tool for project management

Soft Skills

  • Learning about the Agile Method for project management and its framework SCRUM.

Three achievement highlights

  • Playing with BeautifulSoup and Scrapy to scrap data.
  • Learning from a tutorial how to use Logistic Regression to build a text classifier.
  • Learning the difference between Scrapy, BeautifulSoup and Selenium

tasks completed

  • Workspace prepared ( creating a virtual environment and installing the libraries required).
  • Getting familiar with the concepts of NLP ( Word Embedding, Document similarity…).
  • Learning web scrapping and web crawling and their tools.
  • Learning more about Git and how we can use it.
1 Like